We acknowledge funding from the European Commission of a H2020 MSCA award under proposal number 675743 (ISPIC). We thank Tetje van der Sluis and Iris Pardieck for testing the protocol.

All animal experiments were approved by the Animal Experiments Committee of LUMC and were executed according to the animal experimentation guidelines of the LUMC in compliance with the guidelines of Dutch and European committees.
NOTE: For experimental set-up, C57BL/6 mice were subcutaneously inoculated on the right flank with the murine colon tumor MC38 at a concentration of 0.3 x 106 cells/200 &#181;L of phosphate-buffered saline (PBS). After 10 days, when tumors were palpable, the mice were treated with PD-L1 blocking antibodies (clone MIH-5, 200 &#181;g/mouse, intra-peritoneal injection) or were mock-treated. The tumors were resected 3 days later after PD-L1 injection, processed ex vivo, and analyzed by CyTOF mass cytometry using 38 markers13.
1. Equipment and Software for Data Analysis
NOTE: Use a computer (Windows 7 or newer) and processor I5 at 2.4 GHz or equivalent, installed memory RAM 6 GB, and 10 GB of free hard drive space. The R package Cytofast uses existing functions: mainly flowCore, pheatmap, and ggplot. The command lines to be executed in R are included in the protocol. The resource for R instructions can be found at <a href="https://education.rstudio.com/" target="_blank">https://education.rstudio.com/</a>.
<ol>
	<li>To install the Cytofast package, start R (version &#34;3.6&#34;) and install Bioconductor version 3.9 by entering the following code: 
	 
	if (!requireNamespace(&#34;BiocManager&#34;, quietly = TRUE)) 
	install.packages(&#34;BiocManager&#34;) 
	BiocManager::install(&#34;cytofast&#34;)</li>
	<li>Ensure that the package is loaded in the desired environment by running the following: 
	 
	library(cytofast)</li>
</ol>
2. Creating Clusters
NOTE: To showcase the two clustering methods Cytosplore and FlowSOM with Cytofast, the NK cells (CD161+) in the tumor micro-environment 3 days after PD-L1 treatment are analyzed.
<ol>
	<li>Clustering performed by Cytosplore
	<ol>
		<li>After downloading and installing Cytosplore, hosted at &#60;www.Cytosplore.org&#62;, upload the .fcs files (Supplementary Files 1.1&#8211;1.8 [Cytosplore input files]) by clicking File | Open FCS File(s) in Cytosplore. Add a unique sample tag as channel by clicking Add unique sample tag as channel when prompted and select a cofactor for hyperbolic arcsinh transformation (default is 5).</li>
		<li>Select Run H-SNE, run a HSNE level of 3, and wait for the map to be generated. 
		NOTE: This step may require some time, depending on the number of cells analyzed and level of HSNE chosen.</li>
		<li>On the first HSNE level, check the cells that are positive for CD161. Select the CD161+ cells and right-click Zoom into selection. At the second level, repeat the procedure to reach the third level with only CD161+ events.</li>
		<li>Once the last tSNE map is generated, save the clusters defined by Cytosplore by right-clicking on the tSNE map and choose Save Clusters. Choose the directory of the output files as prompted by Cytosplore and note this location, because this directory will then be used to load the .fcs files into R. 
		NOTE: The number of subsets can be manually changed by changing the sigma value. The sigma value is set by default at 30; however, the actual number of subsets depends on the input. Here, Cytosplore detected 10 different subsets, with each file representing one subset.</li>
		<li>Use a simple name (characters only) when renaming the output files, which will make identification and further handling easier. Save the output files by selecting Save. 
		NOTE: After saving, a folder is being created by Cytosplore with the .fcs files, with each file corresponding to the identified clusters in Cytosplore. The next step will be loading the files into R with the help of Cytofast. Here, the generated output files are provided in Supplementary Files 2.1&#8211;2.10 (Cytosplore output files).</li>
		<li>Load the output files generated by Cytosplore into R with the designated function: readCytosploreFCS. 
		 
		dirFCS &#60;- &#34;C:\\Users\\username\\Desktop\\tostudy&#34; 
		cfData &#60;- readCytosploreFCS(dir = dirFCS, colNames = &#34;description&#34;)</li>
		<li>Clean the data by removing some parameters such as &#34;Time&#34; and &#34;Background&#34;. Check the position of the column related to its unnecessary parameters and remove it from the matrix. 
		 
		colnames(cfData@expr) 
		cfData@expr &#60;- cfData@expr[,-c(3,4,6,8:10,46:49,51:54)] 
		 
		NOTE: The unnecessary columns can be seen by reading the column names of the generated matrix. By running colnames(cfData@expr) once more, ensure that only the desired parameters are obtained.</li>
		<li>Re-order the markers so that lineage markers are displayed first, followed by functional markers. 
		 
		cfData@expr &#60;- cfData@expr[,c(1,2,3,35,36,31,9,10,18,8,37,20, 
		29,40,5,30,33,11,34,14,19, 
		32,28,6,7,4,12,13,17,16,15, 
		21,22,24,25,26,27,38,39)] 
		 
		NOTE: Step 2.1.8 is optional.</li>
		<li>Link the meta datafile to the generated data from Cytosplore by uploading the spreadsheet metafile containing clinical information (Supplementary File 3). 
		 
		library(readxl) 
		meta &#60;- read_excel(&#34;C:\\Users\\username\\Desktop\\sample_id.xlsx&#34;) 
		cfData@samples &#60;- data.frame(meta) 
		 
		NOTE: The clustering being performed by Cytosplore is now finished. An alternative clustering option to Cytosplore is FlowSOM and is described in section 2.2. Once one of the two clustering steps is performed, proceed with the visualization step (section 3).</li>
	</ol>
	</li>
	<li>Clustering performed by FlowSOM
	<ol>
		<li>First install FlowSOM in R by running the following command: 
		 
		if (!requireNamespace(&#34;BiocManager&#34;, quietly = TRUE)) 
		install.packages(&#34;BiocManager&#34;) 
		BiocManager::install(&#34;FlowSOM&#34;) 
		library(FlowSOM)</li>
		<li>Install the flowCore package using a similar method and load it in the environment by running the following: 
		 
		if (!requireNamespace(&#34;BiocManager&#34;, quietly = TRUE)) 
		install.packages(&#34;BiocManager&#34;) 
		BiocManager::install(&#34;flowCore&#34;) 
		library(FlowSOM)</li>
		<li>Load the raw data provided in Supplementary Files 4.1&#8211;4.8 (FCS FlowSOM input), which were previously gated on CD161+ events, in R with the read.flowSet function. 
		 
		fcs_raw &#60;- read.flowSet(path=&#34;C:\\Users\\username\\Desktop\\tocluster&#34;, pattern = &#34;.fcs&#34;, transformation = FALSE, truncate_max_range = FALSE, seed=123)</li>
		<li>Select the relevant biological markers (remove &#34;Background&#34; or &#34;Time&#34;) by selecting the proper columns and transform the data in an arcsinh5 manner as shown in the code below (here, remove columns 1, 2, 4, 5, 6, 17, 21, 24, 25, 34, 35, 37, 38, 51, which do not correspond to any biological marker). Apply a cofactor of 5 as previously applied with Cytosplore, by choosing cofactor=5 in the function below. 
		 
		fcs_raw &#60;- fsApply(fcs_raw, function(x, cofactor=5){ 
		colnames(x) &#60;- fcs_raw[[1]]@parameters@data$desc 
		expr &#60;- exprs(x) 
		expr &#60;- asinh(expr[,-c(1,2,4,5,6,17,21,24,25,34,35,37,38,51)]/ cofactor) 
		exprs(x) &#60;- expr 
		return(x)})</li>
		<li>Cluster the data using the FlowSOM function. To compare FlowSOM and Cytosplore, choose to cluster the data in ten subsets as the output previously yielded by Cytosplore. 
		 
		fsom &#60;- FlowSOM(fcs_raw, transformFunction = FALSE, scale = FALSE, 
		scaled.center = FALSE, scaled.scale = FALSE, silent = FALSE, colsToUse = c(1:37), 
		nClus = 10, maxMeta = 10, importance = NULL, seed = 123) 
		 
		NOTE: This can be changed manually by the user.</li>
		<li>Assign each cell to its identified subset and sample ID. 
		 
		subset_id &#60;- as.factor(fsom$FlowSOM$map$mapping[,1]) 
		levels(subset_id) &#60;- fsom$metaclustering 
		head(subset_id)</li>
		<li>Load the metadata file in R (available in Supplementary File 5) containing the group assignment and link it to the .fcs files. 
		 
		sampleid &#60;- read_excel(&#34;C:\\Users\\username\\Desktop\\sample_id.xlsx&#34;) 
		sampleid &#60;- na.omit(sampleid) 
		 
		sampleid$sampleID &#60;- as.factor(sampleid$sampleID) 
		sampleid$group &#60;- as.factor(sampleid$group) 
		sampleid$CSPLR_ST &#60;- as.factor(sampleid$CSPLR_ST) 
		 
		sampleid &#60;- as.data.frame(sampleid) 
		names(sampleid)[3] &#60;- &#34;sampleID&#34; 
		sampleID &#60;- lapply(fsom$FlowSOM$metaData, function(x){rep(x[1], each = length(x[1]:x[2]))}) 
		attr(sampleID, &#39;names&#39;) &#60;- NULL 
		sampleID &#60;- as.factor(unlist(sampleID)) 
		sampleid &#60;- data.frame(sampleid) 
		levels(sampleID) &#60;- paste(&#34;ID&#34;, 1:dim(sampleid)[1], sep=&#34;_&#34;) 
		 
		df &#60;- data.frame(subset_id, sampleID, fsom$FlowSOM$data[, c(1:37)]) 
		rename &#60;- data.frame(colpar=fcs_raw[[1]]@parameters@data$desc) 
		colnames(df) &#60;- c(&#34;clusterID&#34;, &#34;sampleID&#34;, rename$colpar[c(1:37)]) 
		df$ clusterID &#60;- as.factor(df$ clusterID) 
		df$sampleID &#60;- as.factor(df$sampleID)</li>
		<li>Create a cfList based on the dataframe obtained from FlowSOM by running the following script: 
		 
		cfData &#60;- cfList(samples = sampleid, 
		expr = df)</li>
		<li>Re-order the markers to appear similarly to the output from the Cytosplore analysis. 
		 
		cfData@expr &#60;- cfData@expr[,c(1,2,34,36,37,10,23,24,31,22, 
		38,15,8,3,27,9,11,28,35,26, 
		14,33,17,20,21,18,25,29,13, 
		30,12,16,32,4,5,6,7,19,39)] 
		 
		NOTE: The clustering by FlowSOM is now finished. Next, perform visualization of the clustering output.</li>
	</ol>
	</li>
</ol>
3. Visualization: Post-processing Clustering Analysis
NOTE: This step is a method that is common to both clustering methods. Therefore, it can be performed after clustering either with FlowSOM or Cytosplore.
<ol>
	<li>Before creating the heatmaps, generate the count table per sample by using the function cellCounts as shown in the code below. Since some clusters contain fewer cells than others, scale the data per cluster by specifying &#34;scale = TRUE&#34; inside the function cellCounts, so that the dispersion among samples can be easily seen. 
	 
	cfData &#60;- cellCounts(cfData, frequency = TRUE, scale = TRUE) 
	 
	NOTE: The data can now be visualized.</li>
	<li>Visualize with heatmap. 
	NOTE: One of the main functions of this package is cytoHeatmaps, which is used to visualize the phenotype of the created clusters as well as their heterogeneity with respect to the samples. 
	 
	cytoHeatmaps (cfData, group=&#34;group&#34;, legend=TRUE)</li>
	<li>Visualization with boxplots 
	NOTE: Data can be represented in a quantitative way by calling the function cytoBoxplots. The output of this function represents the proportion of each sample for every cluster.
	<ol>
		<li>Generate the cell count as done in step 3.1, but do not scale the data to obtain the frequency of each cluster. 
		 
		cfData &#60;- cellCounts(cfData, frequency = TRUE, scale = FALSE) 
		cytoBoxplots(cfData, group = &#34;group&#34;)</li>
	</ol>
	</li>
	<li>Visualization with median intensity signal histogram 
	NOTE: The data can also be visualized by representing the median intensity signal histogram.
	<ol>
		<li>Visualize the expression intensity of three markers: CD45, CD11c, and CD54.</li>
		<li>Check the names of the desired markers by calling the following line. Note the marker names and include it in the msiPlot function. 
		 
		names(cfData@expr) 
		 
		NOTE: Here, focus on CD45, CD11c, and CD54. Check the exact spelling of the markers and adjust if needed: 
		 
		msiPlot(cfData, markers = c(&#34;89Y_CD45&#34;, &#34;167Er_CD11c&#34;,&#34;164Dy_CD54&#34;), byGroup=&#39;group&#39;)</li>
	</ol>
	</li>
</ol>

<ol>
	<li>Anchang, B., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Visualization+and+cellular+hierarchy+inference+of+single-cell+data+using+SPADE.">Visualization and cellular hierarchy inference of single-cell data using SPADE.</a> Nature Protocols. 11 (7), 1264-1279 (2016).</li><li>Zunder, E. R., Lujan, E., Goltsev, Y., Wernig, M., Nolan, G. P. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=A+continuous+molecular+roadmap+to+iPSC+reprogramming+through+progression+analysis+of+single-cell+mass+cytometry.">A continuous molecular roadmap to iPSC reprogramming through progression analysis of single-cell mass cytometry.</a> Cell Stem Cell. 16 (3), 323-337 (2015).</li><li>Van Gassen, S., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=FlowSOM:+Using+self-organizing+maps+for+visualization+and+interpretation+of+cytometry+data.">FlowSOM: Using self-organizing maps for visualization and interpretation of cytometry data.</a> Cytometry A. 87 (7), 636-645 (2015).</li><li>Levine, J. H., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Data-Driven+Phenotypic+Dissection+of+AML+Reveals+Progenitor-like+Cells+that+Correlate+with+Prognosis.">Data-Driven Phenotypic Dissection of AML Reveals Progenitor-like Cells that Correlate with Prognosis.</a> Cell. 162 (1), 184-197 (2015).</li><li>Samusik, N., Good, Z., Spitzer, M. H., Davis, K. L., Nolan, G. P. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Automated+mapping+of+phenotype+space+with+single-cell+data.">Automated mapping of phenotype space with single-cell data.</a> Nature Methods. 13 (6), 493-496 (2016).</li><li>Spitzer, M. H., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=An+interactive+reference+framework+for+modeling+a+dynamic+immune+system.">An interactive reference framework for modeling a dynamic immune system.</a> Science. 349 (6244), 1259425 (2015).</li><li>Hotelling, H. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Analysis+of+a+complex+of+statistical+variables+into+principal+components.">Analysis of a complex of statistical variables into principal components.</a> Journal of Educational Psychology. 24 (6), 417-441 (1933).</li><li>van der Maaten, L., Hinton, G. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Visualizing+Data+using+t-SNE.">Visualizing Data using t-SNE.</a> Journal of Machine Learning Research. , (2008).</li><li>Pezzotti, N., Hollt, T., Lelieveldt, B., Eisemann, E., Vilanova, A. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Hierarchical+Stochastic+Neighbor+Embedding.">Hierarchical Stochastic Neighbor Embedding.</a> Computer Graphics Forum. 35 (3), 21-30 (2016).</li><li>Becht, E., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Dimensionality+reduction+for+visualizing+single-cell+data+using+UMAP.">Dimensionality reduction for visualizing single-cell data using UMAP.</a> Nature Biotechnology. 37, 38 (2018).</li><li>Haghverdi, L., Buettner, F., Theis, F. J. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Diffusion+maps+for+high-dimensional+single-cell+analysis+of+differentiation+data.">Diffusion maps for high-dimensional single-cell analysis of differentiation data.</a> Bioinformatics. 31 (18), 2989-2998 (2015).</li><li>Beyrend, G., Stam, K., H&#246;llt, T., Ossendorp, F., Arens, R. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Cytofast:+A+workflow+for+visual+and+quantitative+analysis+of+flow+and+mass+cytometry+data+to+discover+immune+signatures+and+correlations.">Cytofast: A workflow for visual and quantitative analysis of flow and mass cytometry data to discover immune signatures and correlations.</a> Computational and Structural Biotechnology Journal. 16, 435-442 (2018).</li><li>Beyrend, G., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=PD-L1+blockade+engages+tumor-infiltrating+lymphocytes+to+co-express+targetable+activating+and+inhibitory+receptors.">PD-L1 blockade engages tumor-infiltrating lymphocytes to co-express targetable activating and inhibitory receptors.</a> Journal for ImmunoTherapy of Cancer. 7 (1), 217 (2019).</li><li>Goeman, J. J., van de Geer, S. A., de Kort, F., van Houwelingen, H. C. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=A+global+test+for+groups+of+genes:+testing+association+with+a+clinical+outcome.">A global test for groups of genes: testing association with a clinical outcome.</a> Bioinformatics. 20 (1), 93-99 (2004).</li></ol>

The authors have nothing to disclose.

Cytofast is a rapid computational tool that provides a quick and global exploration of cytometric data by highlighting and quantifying treatment-specific cellular subsets. The protocol described aims to further process clustering analyses with Cytosplore or FlowSOM. Other clustering analysis tools are suitable for Cytofast, but this requires the use of Cytofast to assign each cell to a subset. Cytofast, however, is not a clustering method, and therefore requires clustering procedures before use.
The analysis performed here showed that certain CD161+ NK cell subsets in the tumor microenvironment were sensitive to a PD-L1 blockade. This was evidenced by changes in their phenotype and abundance, which were observed using both Cytosplore and FlowSOM as clustering methods. Both methods distinguished the main NK cell cluster (CD11b+ NKG2A+) with slightly different frequencies (15%&#8211;20% for Cytosplore, 30%&#8211;40% for FlowSOM). The differences in abundance and this approximation did not affect the global pattern, because both dendrograms displayed in the right panels of Figure 2 and Figure 3 showed similar results. By using Cytofast, it is thus possible (independent from the clustering method chosen) to segregate PD-L1-treated and untreated mice based on analyses of NK cell cluster phenotype and abundance.
Depending on the recorded parameters, modifications to the protocol are needed. Specifically, certain parameters such as time and background must be removed while performing the clustering analysis. In addition, it is important that each cell is assigned to a subset. The cfData function will simply add the raw cell counts per cluster per sample into the cfList. From this step, the cytoheatmap can be built as explained in section 3.
Cytofast has been successfully used as a visualization and quantification tool to compare different clustering methods13. This R package is also compatible with advanced features, such as the globaltest14, which can test associations between groups of clusters using clinical variables. In the future, the globaltest tool and other algorithms can be integrated with Cytofast for more in-depth visualization and quantification.

Mass cytometry (cytometry by time-of-flight, or CyTOF) allows detection of a wide range of intracellular or extracellular biomarkers in millions of single cells. The high-dimensional nature of mass cytometry data necessitates certain analysis tools, such as cell clustering techniques like SPADE1, FlowMaps2, FlowSOM3, Phenograph4, VorteX5, and scaffold maps6. In addition, various dimensionality reduction-based techniques have been developed (i.e., principal component analysis [PCA]7, t-distributed stochastic neighbor embedding [t-SNE]8, hierarchical stochastic neighbor embedding [HSNE]9, uniform manifold approximation and projection [UMAP]10, and diffusion maps11) to improve the speed, interpretation and visualization of high-dimensional datasets.
Downstream analysis of high-dimensional flow and mass cytometric data often lacks automatic processes to perform statistical tests on cluster frequency and links with clinical outcomes. Previously, we developed an R-based workflow known as Cytofast12, which allows for visual and quantitative downstream analyses of clustering techniques by Cytosplore or FlowSOM.
The protocol described here clarifies the use of Cytofast in R and shows how to generate quantitative and qualitative heatmaps and graphs. Furthermore, it facilitates the determination of connections between observed immune phenotypes and clinical outcomes. This report also describes the analysis of a specific mass cytometry dataset using two different clustering procedures: FlowSOM and Cytosplore. By using Cytofast with both clustering methods, it is correspondingly shown that the activation phenotype of NK cells is influenced by PD-L1 immune checkpoint blockade.

<table><tbody><tr><td>Computer</td><td>Dell</td><td>NA</td><td>NA</td></tr></tbody></table>

visualization and quantification of high-dimensional cytometry data using cytofast and the upstream clustering methods flowsom and cytosplore

The complexity of data generated by mass cytometry has necessitated new tools to rapidly visualize analytic outcomes. Clustering methods like Cytosplore or FlowSOM are used for the visualization and identification of cell clusters. For downstream analysis, a newly developed R package, Cytofast, can generate a rapid visualization of results from clustering methods. Cytofast takes into account the phenotypic characterization of cell clusters, calculates the cell cluster abundance, then quantitatively compares groups. This protocol explains the applications of Cytofast to the use of mass cytometry data based on modulation of the immune system in the tumor microenvironment (i.e., the natural killer [NK] cell response) upon tumor challenge followed by immunotherapy (PD-L1 blockade). Demonstration of the usefulness of Cytofast with FlowSOM and Cytosplore is shown. Cytofast rapidly generates visual representations of group-related immune cell clusters and correlations with immune system composition. Differences are observed in the clustering analysis, but separation between groups are visible with both clustering methods. Cytofast visually shows the patterns induced by PD-L1 treatment that include a higher abundance of activated NK cell subsets, expressing a higher intensity of activation markers (i.e., CD54 or CD11c).

The workflow Cytofast (Figure 1) is meant to provide a quantitative and qualitative overview of the data originally clustered by analysis software (i.e., FlowSOM or Cytosplore). Cytofast runs several possible outputs, including the heatmap of all clusters identified in the analysis and based on marker expression (Figure 2 and Figure 3). The dendrogram on the top represents the hierarchical similarity between the identified clusters. The upper panel displays another heatmap showing the relative quantity of corresponding subsets in each sample. The dendrogram on the right shows the similarity between samples and is based on hierarchical clustering performed on the Euclidean distances between samples. The combined heatmaps are shown for FlowSOM followed by Cytofast in Figure 2 and for Cytosplore followed by Cytofast in Figure 3. Cytofast can also be used to present the data quantitatively and display the results in boxplots (by using cytoBoxplots function), as shown in Figure 4 and Figure 5.
Similar clusters were found between the two different methods (e.g., cluster 8 from Cytosplore corresponds to cluster 10 from FlowSOM), and co-expression of some inhibitory markers like PD-1 and LAG-3 were still visible in both methods). Both clustering methods allowed discrimination between PD-L1 vs. PBS treated mice. In contrast, some differences between both methods can be highlighted. FlowSOM identifies 2 clusters (MHC-II+), whereas Cytosplore shows only one cluster (MHC-II+dim). This is due to the initial gating strategy in which NK cells were manually gated on CD161+ cells, then further processed by FlowSOM. However, Cytosplore automatically gated cells from the CD45+ population on the first HSNE level, which were then clustered in a higher hierarchical level. Thus, Cytosplore defined the NK cell subsets more precisely than how manual gating focused on CD161. Nevertheless, hierarchical clustering of the samples was preserved, as shown in the dendrogram on the right, indicating that segregation between the two groups (PD-L1 and PBS) was not dependent on the chosen clustering method.
The number of clusters can be manually defined using both methods. Cytofast enables the user to assess the heterogeneity of their data and can provide insight into how to choose the number of clusters into which the data should be divided. Other features are included in the Cytofast package, such as the msiPlot function (step 3.4.2), showing the median signal Intensity (MSI) plot of every marker per group (Figure 6 and Figure 7). This function allows detection of global changes, such as increases in the expression of CD54 or CD11c in NK cells of the PD-L1-treated group. Optional features can be incorporated in the Cytofast package, such as displaying data in bar graphs and other methods of data representation. The latter requires the addition of ggplot tools, which can be generated by R.
<img alt="Figure 1" class="xfigimg" src="/files/ftp_upload/60525/60525fig01.jpg" /> 
Figure 1: Workflow of Cytofast package. The data were generated by mass cytometry from a tumor 3 days after treatment with immunotherapy or left untreated. Two different clustering techniques were compared: Cytosplore and FlowSOM. Cytofast was used to visualize differences between the two techniques. <a href="https://www.jove.com/files/ftp_upload/60525/60525fig1large.jpg" target="_blank">Please click here to view a larger version of this figure.</a>
<img alt="Figure 2" class="xfigimg" src="/files/ftp_upload/60525/60525fig02.jpg" /> 
Figure 2: Cluster overview and cluster abundance per group as analyzed by Cytofast following Cytosplore. Heatmap of all NK cell clusters (CD161+ cells defined automatically by Cytosplore), which were identified 3 days after immunotherapy (PD-L1). Data shown is based on Cytosplore clustering and pooled from the untreated and PD-L1 treated groups. Levels of ArcSinh5-transformed expression marker are displayed on a rainbow scale. On the lower panel, the relative abundance of each sample is represented by the green-to-purple scale. The dendrogram on the right represents the similarity between samples based on subset frequencies. The frequency scale represents the dispersion of the mean. A low or a high frequency is represented by a green or purple color, respectively. <a href="https://www.jove.com/files/ftp_upload/60525/60525fig02large.jpg" target="_blank">Please click here to view a larger version of this figure.</a>
<img alt="Figure 3" class="xfigimg" src="/files/ftp_upload/60525/60525fig03.jpg" /> 
Figure 3: Cluster overview and cluster abundance per group as analyzed by Cytofast following FlowSOM. Heatmap of all NK cell clusters (pre-gated on CD161+ events), which were identified 3 days after immunotherapy (PD-L1). Data shown is based on FlowSOM clustering and pooled from the untreated and PD-L1 treated groups. Levels of ArcSinh5-transformed expression marker are displayed on a rainbow scale. On the lower panel, the relative abundance of each sample is represented by the green-to-purple scale. The dendrogram on the right represents the similarity between samples based on subset frequencies. The frequency scale represents the dispersion of the mean. A low or a high frequency is represented by a green or purple color, respectively. <a href="https://www.jove.com/files/ftp_upload/60525/60525fig03large.jpg" target="_blank">Please click here to view a larger version of this figure.</a>
<img alt="Figure 4" class="xfigimg" src="/files/ftp_upload/60525/60525fig04.jpg" /> 
Figure 4: Cytofast representation with boxplots of the clusters defined by Cytosplore. The frequency of each cluster is represented in a boxplot, separated into the two groups (PBS and PD-L1). One individual dot corresponds to one mouse. <a href="https://www.jove.com/files/ftp_upload/60525/60525fig4large.jpg" target="_blank">Please click here to view a larger version of this figure.</a>
<img alt="Figure 5" class="xfigimg" src="/files/ftp_upload/60525/60525fig05.jpg" /> 
Figure 5: Cytofast representation with boxplots of the clusters defined by FlowSOM. The frequency of each cluster is represented in a boxplot, separated into the two groups (PBS and PD-L1). One individual dot corresponds to one mouse. <a href="https://www.jove.com/files/ftp_upload/60525/60525fig5large.jpg" target="_blank">Please click here to view a larger version of this figure.</a>
<img alt="Figure 6" class="xfigimg" src="/files/ftp_upload/60525/60525fig06.jpg" /> 
Figure 6: Distribution of signal intensity plots from NK cells automatically gated by Cytosplore. Distribution of signal intensities are shown in a histogram for three specific markers: CD45, CD11c, and CD54. <a href="https://www.jove.com/files/ftp_upload/60525/60525fig6large.jpg" target="_blank">Please click here to view a larger version of this figure.</a>
<img alt="Figure 7" class="xfigimg" src="/files/ftp_upload/60525/60525fig07.jpg" /> 
Figure 7: Distribution of signal intensity plots from NK cells automatically gated by FlowSOM. Distribution of signal intensities are shown in a histogram for three specific markers: CD45, CD11c, and CD54, segregated by groups PBS and PD-L1. <a href="https://www.jove.com/files/ftp_upload/60525/60525fig7large.jpg" target="_blank">Please click here to view a larger version of this figure.</a>
Supplementary Files 1.1&#8211;1.8. <a href="https://cloudflare.jove.com/files/ftp_upload/60525/Supplementary_Files_1.1-1.8.zip" target="_blank">Please click here to view this file (Right click to download).</a>
Supplementary Files 2.1&#8211;2.10. <a href="https://cloudflare.jove.com/files/ftp_upload/60525/Supplementary_Files_2.1-2.10.zip" target="_blank">Please click here to view this file (Right click to download).</a>
Supplementary File 3. <a href="https://cloudflare.jove.com/files/ftp_upload/60525/Supplementary_File_3_sample_id_Cytosplore.xlsx.zip" target="_blank">Please click here to view this file (Right click to download).</a>
Supplementary Files 4.1&#8211;4.8. <a href="https://cloudflare.jove.com/files/ftp_upload/60525/Supplementary_Files_4.1-4.8.zip" target="_blank">Please click here to view this file (Right click to download).</a>
Supplementary File 5. <a href="https://cloudflare.jove.com/files/ftp_upload/60525/Supplementary_File_5._sample_id_FlowSOM.xlsx.zip" target="_blank">Please click here to view this file (Right click to download).</a>

Watch this Scientific Journal Video about Visualization and Quantification of High-Dimensional Cytometry Data using Cytofast and the Upstream Clustering Methods FlowSOM and Cytosplore at JoVE.com

Visualization and Quantification of High-Dimensional Cytometry Data using Cytofast and the Upstream Clustering Methods FlowSOM and Cytosplore

Cytofast is a visualization tool used to analyze output from clustering. Cytofast can be used to compare two clustering methods: FlowSOM and Cytosplore. Cytofast can rapidly generate a quantitative and qualitative overview of mass cytometry data and highlight the main differences between different clustering algorithms.

The complexity of data generated by mass cytometry has necessitated new tools to rapidly visualize analytic outcomes. Clustering methods like Cytosplore ...

visualization-quantification-high-dimensional-cytometry-data-using

Research

JoVE Journal

Immunology and Infection

8.4K Views.  Leiden University Medical Center. Cytofast is a visualization tool used to analyze output from clustering. Cytofast can be used to compare two clustering methods: FlowSOM and Cytosplore. Cytofast can rapidly generate a quantitative and qualitative overview of mass cytometry data and highlight the main differences between different clustering algorithms.

Video: Visualization and Quantification of High-Dimensional Cytometry Data using Cytofast and the Upstream Clustering Methods FlowSOM and Cytosplore

Characterizing Microbiome Dynamics &#8211; Flow Cytometry Based Workflows from Pure Cultures to Natural Communities

The investigation of pure cultures and monitoring of microbial community dynamics is vital to understand and control natural ecosystems and technical applications driven by microorganisms. Next generation sequencing methods are widely utilized to resolve microbiomes, but they are generally resource and time intensive and deliver mostly qualitative information. Flow cytometric microbiome analysis does not suffer from those disadvantages and can provide relative subcommunity abundances and absolute cell numbers at-line. Although it does not deliver direct phylogenetic information, it can enhance the analysis depth and resolution of sequencing approaches. In sharp contrast to medical applications in both research and routine settings, flow cytometry is still not widely used for microbiome analysis. Missing information on sample preparation and data analysis pipelines may create an entry barrier for the researchers facing microbiome analysis challenges that would often be textbook flow cytometry applications. Here, we present three comprehensive workflows for pure cultures, complex communities in clear medium and complex communities in challenging matrices, respectively. We describe individual sampling and fixation procedures and optimized staining protocols for the respective sample sets. We elaborate the cytometric analysis with a complex research centered and an application focused bench top device, describe the cell sorting procedure and suggest data analysis packages. We furthermore propose important experimental controls and apply the presented workflows to the respective sample sets.

Flow cytometric analysis has proven valuable for investigating pure cultures and monitoring microbial community dynamics. We present three comprehensive workflows, from sampling to data analysis, for pure cultures and complex communities in clear medium as well as in challenging matrices, respectively.

The investigation of pure cultures and monitoring of microbial community dynamics is vital to understand and control natural ecosystems and technical ...

Environment

Microfluidic Imaging Flow Cytometry by Asymmetric-detection Time-stretch Optical Microscopy (ATOM)

Scaling the number of measurable parameters, which allows for multidimensional data analysis and thus higher-confidence statistical results, has been the main trend in the advanced development of flow cytometry. Notably, adding high-resolution imaging capabilities allows for the complex morphological analysis of cellular/sub-cellular structures. This is not possible with standard flow cytometers. However, it is valuable for advancing our knowledge of cellular functions and can benefit life science research, clinical diagnostics, and environmental monitoring. Incorporating imaging capabilities into flow cytometry compromises the assay throughput, primarily due to the limitations on speed and sensitivity in the camera technologies. To overcome this speed or throughput challenge facing imaging flow cytometry while preserving the image quality, asymmetric-detection time-stretch optical microscopy (ATOM) has been demonstrated to enable high-contrast, single-cell imaging with sub-cellular resolution, at an imaging throughput as high as 100,000 cells/s. Based on the imaging concept of conventional time-stretch imaging, which relies on all-optical image encoding and retrieval through the use of ultrafast broadband laser pulses, ATOM further advances imaging performance by enhancing the image contrast of unlabeled/unstained cells. This is achieved by accessing the phase-gradient information of the cells, which is spectrally encoded into single-shot broadband pulses. Hence, ATOM is particularly advantageous in high-throughput measurements of single-cell morphology and texture &#8211; information indicative of cell types, states, and even functions. Ultimately, this could become a powerful imaging flow cytometry platform for the biophysical phenotyping of cells, complementing the current state-of-the-art biochemical-marker-based cellular assay. This work describes a protocol to establish the key modules of an ATOM system (from optical frontend to data processing and visualization backend), as well as the workflow of imaging flow cytometry based on ATOM, using human cells and micro-algae as the examples.

Microfluidic Imaging Flow Cytometry by Asymmetric-detection Time-stretch Optical Microscopy ATOM

This protocol describes the implementation of an asymmetric-detection time-stretch optical microscopy system for single-cell imaging in ultrafast microfluidic flow and its applications in imaging flow cytometry.

Scaling the number of measurable parameters, which allows for multidimensional data analysis and thus higher-confidence statistical results, has been the ...

Engineering

Qualitative and Quantitative Analysis of the Immune Synapse in the Human System Using Imaging Flow Cytometry

The immune synapse is the area of communication between T cells and antigen-presenting cells (APCs). T cells polarize surface receptors and proteins towards the immune synapse to assure a stable binding and signal exchange. Classical confocal, TIRF, or super-resolution microscopy have been used to study the immune synapse. Since these methods require manual image acquisition and time-consuming quantification, the imaging of rare events is challenging. Here, we describe a workflow that enables the morphological analysis of tens of thousands of cells. Immune synapses are induced between primary human T cells in pan-leukocyte preparations and Staphylococcus aureus enterotoxin B (SEB)-loaded Raji cells as APCs. Image acquisition is performed with imaging flow cytometry, also called In-Flow microscopy, which combines features of a flow cytometer and a fluorescence microscope. A complete gating strategy for identifying T cell/APC couples and analyzing the immune synapses is provided. As this workflow allows the analysis of immune synapses in unpurified pan-leukocyte preparations and hence requires only a small volume of blood (i.e., 1 mL), it can be applied to samples from patients. Importantly, several samples can be prepared, measured, and analyzed in parallel.

Here, we describe a complete workflow for the qualitative and quantitative analysis of immune synapses between primary human T cells and antigen-presenting cells. The method is based on imaging flow cytometry, which allows the acquisition and evaluation of several thousand cell images within a relatively short period of time.

The immune synapse is the area of communication between T cells and antigen-presenting cells (APCs). T cells polarize surface receptors and proteins ...

Imaging- and Flow Cytometry-based Analysis of Cell Position and the Cell Cycle in 3D Melanoma Spheroids

Three-dimensional (3D) tumor spheroids are utilized in cancer research as a more accurate model of the in vivo tumor microenvironment, compared to traditional two-dimensional (2D) cell culture. The spheroid model is able to mimic the effects of cell-cell interaction, hypoxia and nutrient deprivation, and drug penetration. One characteristic of this model is the development of a necrotic core, surrounded by a ring of G1 arrested cells, with proliferating cells on the outer layers of the spheroid. Of interest in the cancer field is how different regions of the spheroid respond to drug therapies as well as genetic or environmental manipulation. We describe here the use of the fluorescence ubiquitination cell cycle indicator (FUCCI) system along with cytometry and image analysis using commercial software to characterize the cell cycle status of cells with respect to their position inside melanoma spheroids. These methods may be used to track changes in cell cycle status, gene/protein expression or cell viability in different sub-regions of tumor spheroids over time and under different conditions.

We describe two complementary methods using the fluorescence ubiquitination cell cycle indicator (FUCCI) and image analysis or flow cytometry to identify and isolate cells in the inner G1 arrested and outer proliferating regions of 3D spheroids.

Three-dimensional (3D) tumor spheroids are utilized in cancer research as a more accurate model of the in vivo tumor microenvironment, compared to ...

Medicine

ExCYT: A Graphical User Interface for Streamlining Analysis of High-Dimensional Cytometry Data

With the advent of flow cytometers capable of measuring an increasing number of parameters, scientists continue to develop larger panels to phenotypically explore characteristics of their cellular samples. However, these technological advancements yield high-dimensional data sets that have become increasingly difficult to analyze objectively within traditional manual-based gating programs. In order to better analyze and present data, scientists partner with bioinformaticians with expertise in analyzing high-dimensional data to parse their flow cytometry data. While these methods have been shown to be highly valuable in studying flow cytometry, they have yet to be incorporated in a straightforward and easy-to-use package for scientists who lack computational or programming expertise. To address this need, we have developed ExCYT, a MATLAB-based Graphical User Interface (GUI) that streamlines the analysis of high-dimensional flow cytometry data by implementing commonly employed analytical techniques for high-dimensional data including dimensionality reduction by t-SNE, a variety of automated and manual clustering methods, heatmaps, and novel high-dimensional flow plots. Additionally, ExCYT provides traditional gating options of select populations of interest for further t-SNE and clustering analysis as well as the ability to apply gates directly on t-SNE plots. The software provides the additional advantage of working with either compensated or uncompensated FCS files. In the event that post-acquisition compensation is required, the user can choose to provide the program a directory of single stains and an unstained sample. The program detects positive events in all channels and uses this select data to more objectively calculate the compensation matrix. In summary, ExCYT provides a comprehensive analysis pipeline to take flow cytometry data in the form of FCS files and allow any individual, regardless of computational training, to use the latest algorithmic approaches in understanding their data.

ExCYT is a MATLAB-based Graphical User Interface (GUI) that allows users to analyze their flow cytometry data via commonly employed analytical techniques for high-dimensional data including dimensionality reduction via t-SNE, a variety of automated and manual clustering methods, heatmaps, and novel high-dimensional flow plots.

With the advent of flow cytometers capable of measuring an increasing number of parameters, scientists continue to develop larger panels to phenotypically ...

Multicolor Flow Cytometry-based Quantification of Mitochondria and Lysosomes in T Cells

T cells utilize different metabolic programs to match their functional needs during differentiation and proliferation. Mitochondria are crucial cellular components responsible for supplying cell energy; however, excess mitochondria also produce reactive oxygen species (ROS) that could cause cell death. Therefore, the number of mitochondria must constantly be adjusted to fit the needs of the cells. This dynamic regulation is achieved in part through the function of lysosomes that remove surplus/damaged organelles and macromolecules. Hence, cellular mitochondrial and lysosomal contents are key indicators to evaluate the metabolic adjustment of cells. With the development of probes for organelles, well-characterized lysosome or mitochondria-specific dyes have become available in various formats to label cellular lysosomes and mitochondria. Multicolor flow cytometry is a common tool to profile cell phenotypes, and has the capability to be integrated with other assays. Here, we present a detailed protocol of how to combine organelle-specific dyes with surface markers staining to measure the amount of lysosomes and mitochondria in different T cell populations on a flow cytometer.

This article illustrates a powerful method to quantify mitochondria or lysosomes in living cells. The combination of lysosome- or mitochondria-specific dyes with fluorescently conjugated antibodies against surface markers allows the quantification of these organelles in mixed cell populations, like primary cells harvested from tissue samples, by using multicolor flow cytometry.

T cells utilize different metabolic programs to match their functional needs during differentiation and proliferation. Mitochondria are crucial cellular ...

High-Dimensionality Flow Cytometry for Immune Function Analysis of Dissected Implant Tissues

The success of implanting laboratory-grown tissue or a medical device in an individual is subject to the immune response of the recipient host. Considering an implant as a foreign body, a hostile and dysregulated immune response may result in the rejection of the implant, while a regulated response and regaining of homeostasis can lead to its acceptance. Analyzing the microenvironments of implants dissected out under in vivo or ex vivo&#160;settings can help in understanding the pattern of immune response, which can ultimately help in developing new generations of biomaterials. Flow cytometry is a well-known technique for characterizing immune cells and their subsets based on their cell surface markers. This review describes a protocol based on manual dicing, enzymatic digestion, and filtration through a cell strainer for the isolation of uniform cell suspensions from dissected implant tissue. Further, a multicolor flow cytometry staining protocol has been explained, along with steps for initial cytometer settings to characterize and quantify these isolated cells by flow cytometry.

Isolation of cells from dissected implants and their characterization by flow cytometry can significantly contribute to understanding the pattern of immune response against implants. This paper describes a precise method for the isolation of cells from dissected implants and their staining for flow cytometric analysis.

The success of implanting laboratory-grown tissue or a medical device in an individual is subject to the immune response of the recipient host. ...

Bioengineering

Quality-Controlled Sputum Analysis by Flow Cytometry

Sputum, widely used to study the cellular content and other microenvironmental features to understand the health of the lung, is traditionally analyzed using cytology-based methodologies. Its utility is limited because reading the slides is time-consuming and requires highly specialized personnel. Moreover, extensive debris and the presence of too many squamous epithelial cells (SECs), or cheek cells, often renders a sample inadequate for diagnosis. In contrast, flow cytometry allows for high-throughput phenotyping of cellular populations while simultaneously excluding debris and SECs.
The protocol presented here describes an efficient method to dissociate sputum into a single cell suspension, antibody stain and fix cellular populations, and acquire samples on a flow cytometric platform. A gating strategy that describes the exclusion of debris, dead cells (including SECs) and cell doublets is presented here. Further, this work also explains how to analyze viable, single sputum cells based on a cluster of differentiation (CD)45 positive and negative populations to characterize hematopoietic and epithelial lineage subsets. A quality control measure is also provided by identifying lung-specific macrophages as evidence that a sample is derived from the lung and is not saliva. Finally, it has been demonstrated that this method can be applied to different cytometric platforms by providing sputum profiles from the same patient analyzed on three flow cytometers; Navios EX, LSR II, and Lyric. Furthermore, this protocol can be modified to include additional cellular markers of interest. A method to analyze an entire sputum sample on a flow cytometric platform is presented here that makes sputum amenable for developing high-throughput diagnostics of lung disease.

This protocol describes an efficient method for dissociating sputum into a single cell suspension and the subsequent characterization of cellular subsets on standard flow cytometric platforms.

Sputum, widely used to study the cellular content and other microenvironmental features to understand the health of the lung, is traditionally analyzed ...

Sample Preparation for Mass Cytometry Analysis

Mass cytometry utilizes antibodies conjugated with heavy metal labels, an approach that has greatly increased the number of parameters and opportunities for deep analysis well beyond what is possible with conventional fluorescence-based flow cytometry. As with any new technology, there are critical steps that help ensure the reliable generation of high-quality data. Presented here is an optimized protocol that incorporates multiple techniques for the processing of cell samples for mass cytometry analysis. The methods described here will help the user avoid common pitfalls and achieve consistent results by minimizing variability, which can lead to inaccurate data. To inform experimental design, the rationale behind optional or alternative steps in the protocol and their efficacy in uncovering new findings in the biology of the system being investigated is covered. Lastly, representative data is presented to illustrate expected results from the techniques presented here.

This article describes the collection and processing of samples for mass cytometry analysis.

Mass cytometry utilizes antibodies conjugated with heavy metal labels, an approach that has greatly increased the number of parameters and opportunities ...

Biology

Immunofluorescent Microscopy and Analysis for High-Parameter Tissue Imaging

Generating and Analyzing High-Parameter Histology Images with Histoflow Cytometry

The usage of histology to investigate immune cell diversity in tissue sections such as those derived from the central nervous system (CNS) is critically limited by the number of fluorescent parameters that can be imaged at a single time. Most immune cell subsets have been defined using flow cytometry by using complex combinations of protein markers, often requiring four or more parameters to conclusively identify, which is beyond the capabilities of most conventional microscopes. As flow cytometry dissociates tissues and loses spatial information, there is a need for techniques that can retain spatial information while interrogating the roles of complex cell types. These issues are addressed here by creating a method for expanding the number of fluorescent parameters that can be imaged by collecting the signals of spectrally overlapping fluorophores and using spectral unmixing to separate the signals of each individual fluorophore. These images are then processed using an analysis pipeline to take high-parameter histology images and extract single cells from these images so that the unique fluorescent properties of each cell can be analyzed at a single-cell level. Using flow cytometry-like gating strategies, cells can then be profiled into subsets and mapped back onto the histology sections to not only quantify their abundance, but also establish how they interact with the tissue environment. Overall, the simplicity and potential of using histoflow cytometry to study complex immune populations in histology sections is demonstrated.

Author Spotlight: Enhanced Multiplex Immunofluorescent Microscopy Protocol for Neuroscience Research

Described here is a method that can be used to image five or more fluorescent parameters by immunofluorescent microscopy. An analysis pipeline for extracting single cells from these images and conducting single-cell analysis through flow cytometry-like gating strategies is outlined, which can identify cell subsets in tissue sections.

The usage of histology to investigate immune cell diversity in tissue sections such as those derived from the central nervous system (CNS) is critically ...